In‐Solution Digestion of Proteins for Mass Spectrometry

In‐Solution Digestion of Proteins for Mass Spectrometry

50 mass spectrometry: modified proteins and glycoconjugates [3] [3] In‐Solution Digestion of Proteins for Mass Spectrometry By KATALIN F. MEDZIHRAD...

160KB Sizes 0 Downloads 36 Views

50

mass spectrometry: modified proteins and glycoconjugates

[3]

[3] In‐Solution Digestion of Proteins for Mass Spectrometry By KATALIN F. MEDZIHRADSZKY Abstract

Mass spectrometry (MS) has gradually replaced classical methods as a major tool in protein sequencing and characterization. However, the sample preparation repertoire has not changed very much; it has just been adjusted to the needs of the new analytical method. In this chapter frequently used in‐solution enzymatic digestions and chemical cleavages are reviewed. In addition, some practical recommendations as well as the advantages and shortcomings of the methods are discussed.

Introduction

MALDI and electrospray ionization permit intact molecular weight determination of proteins. However, protein identification, the assignment of covalent modifications, the determination of sequence errors, as well as de novo sequencing are carried out usually at the peptide level. The digestion method and the conditions have to be carefully selected based on the protein sequence as well as on the desired information to be obtained. In‐solution digestion is preferred over in‐gel digestion for more control over the outcome of the process. The conditions (i.e., the pH, protein concentration, digestion buffers, additives, the proteolytic enzyme, and enzyme/substrate ratio) can be altered for in‐solution digestions more easily according to the researchers’ needs, and the recovery of the digestion products is more reliable. Recombinant proteins produced for therapeutical purposes are usually characterized by mass spectrometry (MS) after carefully controlled in‐solution digestion (Bloom et al., 1996; Guzzetta et al., 1993; Ling et al., 1991; Medzihradszky et al., 1997; Rush et al., 1995). Limited proteolysis is carried out frequently when the research is focused on the identification of structural and functional domains of proteins (Ostrelund et al., 1999). Similarly, in‐solution digestion is applied usually when protein/protein interactions are probed. For example, an antigen/antibody complex can be subjected to proteolysis, during which the ‘‘protected’’ portion of the antigen remains intact (Jemmerson and Paterson, 1986; Sheshberadaran and Payne, 1988) and can be separated from the antibody to be characterized by MS (van de Water et al., 1997). METHODS IN ENZYMOLOGY, VOL. 405 Copyright 2005, Elsevier Inc. All rights reserved.

0076-6879/05 $35.00 DOI: 10.1016/S0076-6879(05)05003-2

[3]

in‐solution digestions for MS

51

The idea of ‘‘protected’’ surfaces has been utilized in identifying interactive surfaces of proteins as well as substrate binding pockets. Sequence stretches involved in protein/protein interactions or ‘‘covered’’ with the substrate are not available for H‐D exchange (Mandell et al., 1998, 2001) or chemical derivatization (Everett et al., 1990) and thus can be identified by mass spectrometric characterization following the appropriate proteolysis. Obviously, in order to gain information on the tertiary structure of proteins, one has to retain the native conformation as much as possible. In‐solution covalent labeling or intramolecular cross‐linking of proteins may be then followed by in‐solution digestion and MS characterization of the products (Tschirret‐Guth et al., 1999; Young et al., 2000). This chapter presents an overview of some frequently used in‐solution cleavages, with practical hints, as well as the advantages and shortcomings of some methods listed. Steps Prior to the Digestion

If the final protein purification step is gel‐electrophoresis, electroelution can be used to transfer the protein into solution. However, successful elimination of the sodium dodecyl sulfate (SDS) by acetone precipitation (Konigsberg and Henderson, 1983) as well as by chromatography (Simpson et al., 1987) may be possible although their use certainly leads to significant protein losses. Some researchers recommend a simple elution from the gel, using a formic acid/acetonitrile/isopropanol/water 50:25:15:10 mixture (v/v/ v/v) (Feick and Shiozawa, 1990). In general, the purification protocols have to be altered so that the resulting protein solution will not contain anything that is interfering with the following digestion and cannot be easily removed prior to the analysis steps if necessary. Traditional protocols frequently called for complex buffers with high salt concentrations, for additives to stabilize the enzymes or to slightly increase their activity, and for detergents to disrupt the tertiary structure of the protein to be digested. For digests prepared for analysis by MS, volatile buffers of lower salt concentrations [25–100 mM] are recommended; additives, unless absolutely necessary, are eliminated. To keep the protein soluble and to denature it, urea, guanidine hydrochloride, or organic solvents, such as acetonitrile, are preferable to detergents. Denaturing proteins disrupts their tertiary structure and, thus, may provide access for the digesting enzymes to all sites. However, secreted and membrane proteins usually feature disulfide bridges that will not be disrupted by simple denaturing measures. Disulfide bonds have to be reduced, and to prevent reoxidation, the newly formed free sulfhydryls have to be blocked prior to the digestion. For best results, the protein

52

mass spectrometry: modified proteins and glycoconjugates

[3]

should be dissolved in 6 M guanidine hydrochloride and in a buffer of pH 8.0. (If solid guanidine hydrochloride is added to a solution to achieve a final concentration of 6 M, the volume of the solution will increase 1.75‐ fold!). Approximately 500‐fold excess of dithiothreithol (DTT) at 60 for 1 h will complete the reduction. A 1100‐fold excess of iodoacetic acid sodium salt is then added to the mixture and is incubated at room temperature for 1.5 h, under Ar, in the dark. Iodoacetic acid may be preferred to other derivatizing agents because most proteolytic enzymes are active at mildly basic pH, and the additional acidic residues may make the proteins more soluble under these conditions. Once the derivatization is complete, the reagent excess has to be removed either by high‐performance liquid chromatography (HPLC) or by dialysis. ‘‘Leftover’’ alkylating agents at pH 8 may derivatize the e‐amino groups of Lys residues, the side chains of Met and His residues, as well as the N‐termini of the peptides formed during the digestion. Both modes of purification lead to some protein losses. In general, dialysis is more recommended than HPLC, since for stickier proteins, it can be performed with urea‐ or guanidine hydrochloride‐containing buffers that may prevent protein precipitation. If this removal step is skipped, the reduction/alkylation should be performed in the presence of a denaturing agent tolerated by the endoprotease, and the solution should be diluted accordingly prior to the digestion. However, the recommended protein concentration for digestion is not lower than 25 g/ml. Other alkylating agents, such as iodoacetamide, vinylpyridine, etc., also can be applied. Using volatile reducing and alkylating agents, such as triethylphosphyne and iodoethanol, respectively, may permit the elimination of the desalting step (Hale et al., 2004). Glycosylation of the protein also may hinder its accessibility to proteolytic digestions. Thus, in certain instances, the removal of the N‐linked carbohydrates (i.e., incubation with peptide N‐glycosidase F [PNGase F]) has to be considered. In such cases, one has to keep in mind that PNGase F converts the previously glycosylated Asn residues into aspartic acids. Enzymatic Digestions

A wide variety of endoproteases can be used for protein cleavages. Digestion protocols may have to be altered to eliminate some of the originally used components, salts, and detergents that are not compatible with mass spectrometry as discussed earlier. In addition, for mass spectrometric purposes, enzymes of high specificity are preferred because the masses of the expected products have to be predicted. Each protein requires different conditions for optimal digestion. While bovine fetuin (341 residues, N‐ and O‐glycosylated protein) is fully digested

[3]

in‐solution digestions for MS

53

with 1% (w/w) trypsin in 1 h (Medzihradszky et al., 1994), Factor VIII polypeptides (368, 372, 643, and 909 residues, variably glycosylated) require a much longer digestion time (16 – 20 h) with 4% enzyme (Medzihradszky et al., 1997). The same protein may be readily accessible for certain enzymes, while it may also require the use of denaturing agents for others. The relatively small shark liver fatty‐acid‐binding protein (132 residues) required the presence of 2 M urea for endoprotease Lys‐C or Glu‐C digestions, while it was completely digested with trypsin without denaturation (Medzihradszky et al., 1992). Tables I and II show the cleavage specificity, pH optimum, and recommended digestion conditions for frequently used endoproteases (Allen, 1989; Riviere et al., 1991). In addition, urea, guanidine hydrochloride, and acetonitrile concentrations are listed when the enzyme still retains a significant portion of its activity. However, long incubations in the presence of urea may result in carbamoylation of the available N‐termini (43 Da mass increase) (Wen et al., 1992) as well as the e‐amino groups of Lys‐residues (þ43 Da and prevented tryptic or Lys‐C cleavage). It is important to remember that endoproteases are usually very poor exoproteases. When there are a series of potential cleavage sites in close proximity, once one has been cleaved, the others may be too close to the termini to be digested. Similarly, it has to be considered that most enzymes lose specificity during long incubation times. For example, endoprotease Lys‐C will cleave at most tryptic sites. Trypsin will show more chymotryptic activity, but given enough time, ‘‘tryptic’’ cleavage has been observed at about any amino acid. In addition, one has to consider that these enzymes will digest every protein present, including themselves, in a given mixture. Though endoproteases are added to digestion mixtures, usually in substoichiometric quantities, autolysis products are frequently detected. Side chain‐protected trypsin manufactured by reductive alkylation of the e‐ amino groups of Lys residues was developed to minimize autolysis by blocking some of the cleavage sites (Rice et al., 1977). However, this attempt was not completely successful. Table III shows some CID‐characterized autolysis products of the side‐chain‐modified porcine trypsin (Promega, Madison, WI). Our results, i.e., the observation of N‐terminally and C‐terminally modified peptides (Table III), suggest that some of the autolysis must have happened during the derivatization of the protease. Interestingly, the removal of Lys residues as cleavage sites promotes hydrolysis at the C‐terminus of Asn residues. This phenomenon was also observed and reported for wild‐type trypsin, though at a lesser extent (Vestling et al., 1990). Enzymes can be tricked by chemical modifications: for example, O‐glycosylated, O‐phosphorylated, or sulfated Ser and Thr residues after ‐elimination and Michael addition of 2‐aminoethanethiol

54

TABLE I HIGHLY SPECIFIC ENDOPROTEASES Specificitya

pH

Trypsinc(EC 3.4.21.4)

Arg#, Lys#

7–8.5

Endoprotease Arg‐Ce Clostridium histolycum (EC 3.4.22.8) Endoprotease Glu‐C Staphylococcus aureus V8 (EC 3.4.21.19) Endoprotease Lys‐C Achromobacter lyticus (EC 3.4.21.50) Endoprotease Asp‐N Pseudomonas fragi mutant (EC 3.4.24.33) Prolyl endopeptidasef Flavobacterium meningosepticum (EC 3.4.21.26)

Arg#

7.2–8

Glu#, Asp#

7.5–8.5

Lys#

8–9.5

#Asp, #cysteic acid (#Glu)

6–8.5

Pro#

7–7.5

a

Toleratesb d Urea: 6 M Gu.HCl: 1 M MeCN: 40% Urea: 4 M MeCN: 10% Urea: 2 M Gu.HCl: 1 M MeCN: 20% Urea: 8 M Gu.HCl: 2 M MeCN: 40% Urea: 1 M Gu.HCl: 1 M MeCN: 10%

Recommended (37 ) 1–5%; 2–18 h

0.5–2%; 1–18 h 1–5 mM DTTe 1–5%; 4–18 h

0.5–2%; 2–6 h

0.5–5%; 2–18 h

0.1–1%; 1–4 h

[3]

Certain residues adjacent to the cleavage site may slow down the hydrolysis process. Pro residues usually have a prohibitive affect. This recommendation is based on observations in the authors’ laboratory, on manufacturers’ recommendations, and data published (Riviere et al., 1991). Gu.HCl stands for guanidine hydrochloride; MeCN stands for acetonitrile. c Trypsin preparations always display some chymotryptic activity, and autolysis may yield peptides with C‐terminal Asn residues, especially if the Lys‐cleavage sites are blocked (see Table III). d Long incubation in the presence of urea may result in the carbamoylation of the newly formed N‐termini as well as the e‐amino groups of Lys residues, thus eliminating this residue as tryptic or Lys‐C cleavage site. e Requires the presence of thiol for full activity. f This enzyme does not digest proteins but will cleave peptides approximately up to 20 to 30 residues. b

mass spectrometry: modified proteins and glycoconjugates

Enzyme (IUBMB enzyme nomenclature)

[3]

TABLE II LESS SPECIFIC ENDOPROTEASES Enzyme (IUBMB enzyme nomenclature)

Pepsinc (EC 3.4.23.1)

Proteinase K Tritirachium album (EC 3.4.21.64) Thermolysin (EC 3.4.24.27)

Phe#, Trp#, Tyr#, (Leu#, Met#) (Adjacent Pro prevents cleavage) Phe#, Met#, Leu# (Adjacent hydrophobic residues preferred; prolyl peptide bonds are not cleaved) Aliphatic residue# Aromatic residue# Hydrophobic residue# #Leu, #Ile, #Phe, #Trp, #Met, #Val (Residue cannot have a Pro at its C terminus)

pH 7–9

Toleratesb

Recommended

Urea; 2 M Gu.HCl; 2 M MeCN; 30%

1–5%; 2–18 h, 37

2–4

1%1–2 h, RT

6.5–9.5

Urea: 2 M

1–2%;1–8 h, 37

7–9

Up to 80 Urea: 8 M

0.5–5%; 45 , 2–6 h

in‐solution digestions for MS

Chymotrypsin (EC 3.4.21.1)

Preferencea

a

While chymotrypsin exhibits specificity for aromatic amino acids, the other enzymes cleave rather randomly, displaying some preferences. These recommendations are based on observations in the authors’ laboratory, on manufacturers’ recommendations, and data published (Allen, 1989; Riviere et al., 1991). Gu.HCl stands for guanidine hydrochloride; MeCN stands for acetonitrile. c For quick digestions, immobilized pepsin may be used at a 1:1 ratio for a few minutes (Mandell et al., 1998). b

55

56

MHþcalc.

Position

Sequence

515.3306 842.5100 856.5256 870.5412 1045.5642 1126.5645 1420.7225 1531.8405 1940.9354 2003.0734 2211.1046 2225.1202 2239.1358 2283.1807 2299.1756 2678.3822 2807.3145 2914.5062 3094.6246 3337.7577 3353.7526

[46–49] [100–107] [100–107] [100–107] [90–99] [70–79] [212–223] [84–97] [50–66] [80–97] [50–69] [50–69] [50–69] [70–89] [70–89] [76–99] [20–45] [50–75] [70–97] [70–99] [70–99]

IQVR (R)VATVSLPR(S) (R)V*ATVSLPR(S) (R)V**ATVSLPR(S) (K)LSSPATLNSR(V) (K)IITHPNFNGN(T) (N)YVNWIQQTIAAN(<) (N)DIM(O)LIKLSSPATLN(S) (R)LGEHNIDVLEGNEQFIN(A) (N)TLDNDIM(O)LIK**LSSPATLN(S) (R)LGEHNIDVLEGNEQFINAAK(I) (R)LGEHNIDVLEGNEQFINAAK*(I) (R)LGEHNIDVLEGNEQFINAAK**(I) (K)IITHPNFNGNTLDNDIMLIK(L) (K)IITHPNFNGNTLDNDIM(O)LIK(L) (N)FNGNTLDNDIM(O)LIK**LSSPATLNSR(V) (N)SGSHFC*GGSLINSQWVVSAAHC*YK**SR(I) (R)LGEHNIDVLEGNEQFINAAK**IITHPN(F) (K)IITHPNFNGNTLDNDIMLIK**LSSPATLN(S) (K)IITHPNFNGNTLDNDIMLIK**LSSPATLNSR(V) (K)IITHPNFNGNTLDNDIM(O)LIK**LSSPATLNSR(V)

[3]

Note: N‐terminal amino acids or Lys residues labeled with * or ** indicate the presence of 1 or 2 methylgroups at the terminus or on the side chain, respectively; C* ¼ half cystine; Met(O) ¼ Met sulfoxide. For a more complete list, visit http://prospector.ucsf.edu.

mass spectrometry: modified proteins and glycoconjugates

TABLE III FREQUENTLY OBSERVED AUTOLYSIS PRODUCTS OF SIDE CHAIN‐PROTECTED PORCINE TRYPSIN (PROMEGA), CHARACTERIZED BY LOW ENERGY ESIMS–CID

[3]

in‐solution digestions for MS

57

will become Lys analogues and, thus, cleavage sites for trypsin or endoprotease Lys‐C (Rusnak et al., 2002). The specificity of recombinant enzymes may be altered by design: for example, a Tyr‐specific trypsin mutant has been described (Pal et al., 2004).

Immobilized Enzymes

A variation for in‐solution digestion is the utilization of immobilized enzymes. Some enzymes that are immobilized on HPLC cartridges are commercially available; these include trypsin, endoprotease Glu‐C, and pepsin (e.g., from Pierce, Rockford, IL). The cartridges are reusable. Small volume samples can be digested on‐column with little dilution. The high enzyme excess that can be applied accelerates the digestion process: the reaction time can be controlled by the flow rate. Autolysis products can be eliminated this way. The digestion and the following fractionation readily can be automated (Hara et al., 2000; Hsieh et al., 1996). The determination of in vivo adducts of mitochondrial aldehyde dehydrogenase with disulfiram, a drug used in the aversion therapy treatment of alcoholics, is an excellent demonstration for the combination of on‐line in‐solution digestion and mass spectrometry (Shen et al., 2001). Immobilized enzymes are preferred when studying H‐D‐exchanged proteins. To prevent the loss of the labels, the proteolysis as well as MS analysis has to be accelerated. These studies are carried out performing digestions at 0 with immobilized pepsin (1:1 w/w) that reduces the digestion time to approximately 10 min, and the samples are analyzed by MALDI mass spectrometry with ‘‘frozen’’ sample introduction (Mandell et al., 1998). Trypsin and endoprotease Glu‐C also have been used immobilized on paramagnetic beads (Krogh et al., 1999). This approach may be preferable when a series of different enzymes have to be used to achieve the desired results. Microfluidic reactors containing immobilized enzymes also have been fabricated (Krenkova and Foret, 2004).

Chemical Cleavages

Only a handful of chemical methods yield relatively predictable cleavage products; however, these chemicals may succeed when enzymes fail or provide more specific alternative methods for sequences that can be digested only by nonspecific endoproteases (Allen, 1989; Smith, 1997). Prior reduction of the disulfide bridges and alkylation of the sulfhydryls that helps enzymatic digestions also may lead to higher yields for chemical cleavages.

58

mass spectrometry: modified proteins and glycoconjugates

[3]

Cleavage of Asn‐Gly Bonds The cleavage of Asn‐Gly bonds can be performed with 2‐M hydroxylamine solution at pH 9 in the presence of 2‐M guanidine hydrochloride at 45 for 4 h. Since the Gly residue does not have a side‐chain, there is no steric hindrance, and the Asn residue may form a cyclic imide. Its ‐amide reacts with the amide N of the peptide bond between the Asn and Gly residues:

This succinimide reacts with the hydroxylamine, yielding a free N‐ terminus at the Gly‐side and an aspartyl hydroxamate as the C terminus of the other new peptide, which alters its elemental composition and the mass accordingly by an additional O atom (Blodgett et al., 1985; Bornstein and Balian, 1977). Both the ‐NH‐CH(CH2‐COOH)‐CO‐NHOH and ‐ aspartyl hydroxymate NH‐CH(CH2‐CO‐NHOH)‐COOH may form. The reaction can be stopped by acidifying the mixture. With extended reaction time, Asn‐any residue bonds may be cleaved, and Asn and Gln residues may yield hydroxamate derivatives (Bornstein and Balian, 1977). Similarly, some posttranslational modifications may not survive cleavage conditions: for example, fatty acids attached to Cys residues may be released (Weimbs and Stoffel, 1992). Cleavage of Asp‐Xxx Bonds This cleavage is carried out with diluted (10 mM) hydrochloric acid in a sealed tube at 108 for 2 h (Smith, 1997). Under these conditions, a series of side reactions can be expected, such as the deamidation of Gln and Asn residues, their potential cyclization, the peptide bond cleavage (see previous paragraphs), and the decomposition of Trp residues. In case of glycoproteins, acid‐sensitive neuraminic acid will be lost as well. Cleavage of Cys‐Xxx Bonds The Cys residues are converted to S‐cyanocysteine by reaction with 2‐nitro‐5‐thiocyanobenzoate at pH 8. The cyanylated protein is cleaved at the Cys residues by incubation at pH 9 at 37 for 16 h or longer (Jacobson et al., 1983). The new peptides will bear an iminothiazolidine‐4‐carboxyl residue:

[3]

in‐solution digestions for MS

59

at their N‐termini, while the new C‐termini will be free of carboxylic acids. Wu and Watson (1997) modified this protocol for disulfide‐bridge assignment. Partial reduction was carried out under acidic conditions (pH 3) to prevent disulfide shuffling, and newly formed sulfhydryls were immediately cyanylated by 1‐cyano‐4‐dimethylamino‐pyridinium tetrafluoroborate (Wakselman and Guibe‐Jampel, 1976). Extended incubation at high pH may lead to the cleavage of additional peptide bonds as well as to the ‐elimination of O‐linked carbohydrates, H3PO4 or H2SO4, from modified Ser and Thr residues (Medzihradszky et al., 2004; Rusnak et al., 2002). Cleavage of Met‐Xxx Bonds The cleavage is performed with large excess of CNBr in acid at room temperature for 12 to 24 h. The new peptides have a homoserine: ‐NH‐CH(CH2‐CH2‐OH)‐COOH open (residue weight 101.0477) or forming a lacton ring at their C‐termini (Gross and Witkop, 1961). Met‐Thr, and at a lesser extent Met‐Ser, bonds may be cleaved with a lower yield even when the Met is converted to homoserine (Schroeder et al., 1969). Because of the low pH and long incubation times, Asp‐Xxx, especially Asp‐Pro bonds, may be cleaved, and other side reactions listed previously for acidic conditions may occur. Traditionally, the reaction was carried out in 70% formic acid. Incubation with formic acid causes the formylation of hydroxy‐amino acids, the newly formed homoserines included (Beavis and Chait, 1990). This esterification is reversible by incubation with 0.1% trifluoroacetic acid (TFA) in water at room temperature for approximately 24 h. To prevent such side reactions, the formic acid can be replaced by 50 to 70% TFA, which does not produce esterified side chains. Analysis of the Digests

When the analysis has to be accelerated, for example, in H‐D exchange studies, MALDI–TOF analysis of the unseparated digest is the method of choice (Mandell et al., 1998, 2001). Whenever high sequence coverage is a prerequisite for the success in order to achieve this outcome successfully, the digest should be fractionated by reversed‐phase HPLC and subjected to MS analysis on‐line or off‐line.

60

mass spectrometry: modified proteins and glycoconjugates

[3]

Since the presence of TFA adversely affects the detection sensitivity in electrospray ionization, formic acid is recommended as the ion pair‐forming reagent of the mobile phase for on‐line liquid chromatography (LC)–MS analysis. To achieve chromatographic resolution comparable to that with the TFA‐ containing solvents, the acetonitrile may be replaced by an ethanol/propanol 5:2 mixture (Medzihradszky et al., 1994). In addition, when formic acid is the ion‐pairing agent negatively charged molecules, such as sialylated glycopeptides, phospho‐ and sulfo‐peptides will feature longer retention times than their unmodified counterparts (Medzihradszky et al., 1994, 2004). For comprehensive protein characterization, picomoles, if not nanomoles, of the protein should be available, even if we have instrumentation that is routinely capable of femtomole‐level sample detection. One of the best‐documented examples of comprehensive protein characterization is bovine fetuin. From a single LC‐MS analysis of about 20 picomoles of a tryptic digest, reproducibly almost the entire sequence is covered, as shown here:

In addition, from the masses observed, the carbohydrate‐heterogeneity at the N‐glycosylation sites (Asn‐81, Asn‐138, and Asn‐158 [labeled with asterisks]) could be addressed. It can also be determined that one of these sites, Asn‐158, is not 100% occupied. Similarly, the carbohydrate heterogeneity for the O‐linked glycopeptide [228–288] can be addressed (Medzihradszky et al., 1994). However, the structure of carbohydrates and the sites of O‐glycosylation cannot be determined from a single experiment. In addition, another O‐linked glycopeptide, [316–330], which is present at much lower quantities, goes usually undetected without prior enrichment of the O‐linked species by Jacaline‐agarose chromatography (R. R. Townsend and K. F. Medzihradszky, unpublished results). Fetuin also contains phosphopeptides that were discovered only recently (Thompson et al., 2003). This is a common problem: covalent labels, xenobiotic, or posttranslational modifications are usually present in a digest in substoichiometric quantities. In addition, some of them, like small, highly glycosylated peptides, phosphopeptides, or sulfopeptides may be too hydrophilic to be retained on the column during fractionation. Very hydrophobic species, such as palmitoylated peptides or transmembrane

[3]

in‐solution digestions for MS

61

regions, may never elute. The fractionation has to be tailored to the project. Thus, special tracking, such as monitoring a characteristic UV‐ absorbance or radioactivity, could be employed during fractionation if the study is aimed at the characterization of some specific modification. Fragmentation induced in the ion source may permit the identification of modified compounds that produce unique fragment ions. For example, glycopeptide‐containing fractions can be identified by monitoring the presence of a mass‐to‐charge ratio (m/z) 204 ion, an oxonium fragment for N‐ acetylhexosamines (Huddleston et al., 1993), while phosphopeptides and sulfopeptides yield diagnostic negative ion fragments at m/z 79 and 80, respectively (Bean et al., 1995). Precursor ion scanning also may be employed to identify peptides containing a certain residue or modification. Carbohydrate ions may be used for glycopeptide identification (Carr et al., 1993), and immonium ions can be used to identify peptide ions barely above the noise level (Wilm et al., 1996). Scanning for the precursors of the phosphorylated immonium ion of Tyr aids the identification of such modified peptides (Steen et al., 2001). In general, phosphopeptides can be identified as precursors of the m/z 79 negative ion (Carr et al., 1996). Similarly, isotope‐coded affinity tag (ICAT)‐modified peptides (Gygi et al., 1999) can be identified as precursors of diagnostic fragments (Baldwin et al., 2001). However, with proper LC‐MS conditions, most components elute in narrow peaks (i.e., on‐line precursor ion scanning may not be possible at the desired sensitivity level). Thus, precursor ion scanning can be performed most efficiently on nanospray‐introduced HPLC fractions of the digests. Frequently, a special enrichment method can be applied instead of or prior to reversed‐phase chromatography. Such purification methods are, for example, the immobilized metal ion affinity chromatography (IMAC)‐enrichment of phosphopeptides (Ficarro et al., 2002; Neville et al., 1997; Nuwaysir and Stults, 1993; Posewitz and Tempst, 1999; Zhou et al., 2000), the lectin‐based affinity chromatography of sugars (Hortin, 1990; Krogh et al., 1999; Treuheit et al., 1992), as well as the extraction of biotinylated peptides with avidin (Girault et al., 1996). Even the enriched fractions are usually subjected to multiple MS analyses, first to determine the complexity of the mixture and the peptide masses and then to gain more detailed structural information by the MS/MS analysis of the selected components. Summary

Endoproteases or chemical methods can be applied under controlled conditions to produce peptides with predictable results. A wide variety of software programs are available for such predictions, such as MS‐digest

62

mass spectrometry: modified proteins and glycoconjugates

[3]

in ProteinProspector (University of California, San Francisco, http:// prospector.ucsf.edu), PAWS (ProteoMetrics, New York, http://prowl. rockefeller.edu), or Sherpa (Biochemistry Department, University of Washington, http://hairyfatguy.com/Sherpa). In‐solution digestions are recommended when the native conformation of the protein has to be retained as well as when complete sequence coverage is desirable. For the analysis of such digests on‐line LC‐MS with high chromatographic resolution is preferable. For comprehensive protein characterization, multiple digestions and analytical steps may be necessary. Thus, the sample requirement for such analyses may be much higher than the detection sensitivity of one’s mass spectrometer. Acknowledgment This work was supported by NIH grants NCRR RR01614, RR01296, RR014606, and RR015804 to the UCSF Mass Spectrometry Facility Director A. L. Burlingame.

References Allen, G. (1989). Specific cleavage of the protein. In ‘‘Sequencing of Proteins and Peptides,’’ 2nd ed., pp.73–104. Elsevier, Amsterdam. Baldwin, M. A., Medzihradszky, K. F., Lock, C. M., Fisher, B., Settineri, C. A., and Burlingame, A. L. (2001). Matrix‐assisted laser desorption/ionization coupled with quadrupole/orthogonal acceleration time‐of‐flight mass spectrometry for protein discovery, identification, and structural analysis. Anal. Chem. 73, 1707–1720. Bean, M. F., Annan, R. S., Hemling, M. E., Mentzer, M., Huddleston, M. J., and Carr, S. A. (1995). LC–MS methods for the selective detection of post‐translational modifications in proteins: Glycosylation, phosphorylation, sulfation, and acylation. In ‘‘Techniques in Protein Chemistry VI’’ (J. W. Crabb, ed.), pp. 107–116. Academic Press, San Diego, CA. Beavis, R. C., and Chait, B. T. (1990). Rapid, sensitive analysis of protein mixtures by mass spectrometry. Proc. Natl. Acad. Sci. USA 87, 6873–6877. Blodgett, J. K., Londin, G. M., and Collins, K. D. (1985). Specific cleavage of peptides containing an aspartic‐acid (beta‐hydroxamic acid) residue. J. Am. Chem. Soc. 107, 4305–4313. Bloom, J. W., Madanat, M. S., and Ray, M. K. (1996). Cell line and site specific comparative analysis of the N‐linked oligosaccharides on human ICAM‐1des454–532 by electrospray ionization mass spectrometry. Biochemistry 35, 1856–1864. Bornstein, P., and Balian, G. (1977). Cleavage at Asn‐Gly bonds with hydroxylamine. Meth. Enzymol. 47, 132–145. Carr, S. A., Huddleston, M. J., and Bean, M. F. (1993). Selective identification and differentiation of N‐ and O‐linked oligosaccharides in glycoproteins by liquid chromatography–mass spectrometry. Protein Sci. 2, 183–196. Carr, S. A., Huddleston, M. J., and Annan, R. S. (1996). Selective detection and sequencing of phosphopeptides at the femtomole level by mass spectrometry. Anal. Biochem. 239, 180–192. Everett, E. A., Falick, A. M., and Reich, N. O. (1990). Identification of a critical cysteine in EcoRI DNA methyltransferase by mass spectrometry. J. Biol. Chem. 265, 17713–17719.

[3]

in‐solution digestions for MS

63

Feick, R. G., and Shiozawa, J. A. (1990). A high‐yield method for the isolation of hydrophobic proteins and peptides from polyacrylamide gels for protein sequencing. Anal. Biochem. 187, 205–211. Ficarro, S. B., McCleland, M. L., Stukenberg, P. T., Burke, D. J., Ross, M. M., Shabanowitz, J., Hunt, D. F., and White, F. M. (2002). Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat. Biotechnol. 20, 301–305. Girault, S., Chassaing, G., Blais, J. C., Brunot, A., and Bolbach, G. (1996). Coupling of MALDI–TOF mass analysis to the separation of biotinylated peptides by magnetic streptavidin beads. Anal. Chem. 68, 2122–2126. Gross, E., and Witkop, B. (1961). Selective cleavage of the methionyl peptide bonds in ribonuclease with cyanogen bromide. J. Am. Chem. Soc. 83, 1510–1511. Guzzetta, A. W., Basa, L. J., Hancock, W. S., Keyt, B. A., and Bennett, W. F. (1993). Identification of carbohydrate structures in glycoprotein peptide maps by the use of LC–MS with selected ion extraction with special reference to tissue plasminogen activator and a glycosylation variant produced by site directed mutagenesis. Anal. Chem. 65, 2953–2962. Gygi, S. P., Rist, B., Gerber, S. A., Turecek, F., Gelb, M. H., and Aebersold, R. (1999). Quantitative analysis of complex protein mixtures using isotope‐coded affinity tags. Nat. Biotechnol. 17, 994–999. Hale, J. E., Butler, J. P., Gelfanova, V., You, J. S., and Knierman, M. D. (2004). A simplified procedure for the reduction and alkylation of cysteine residues in proteins prior to proteolytic digestion and mass spectral analysis. Anal. Biochem. 333, 174–181. Hara, S., Katta, V., and Lu, H. S. (2000). Peptide map procedure using immobilized protease cartridges in tandem for disulfide linkage identification of neu differentiation factor epidermal growth factor domain. J. Chromatogr. A. 867, 151–160. Hortin, G. L. (1990). Isolation of glycopeptides containing O‐linked oligosaccharides by lectin affinity chromatography on jacalin‐agarose. Anal. Biochem. 191, 262–267. Hsieh, Y. L. F., Wang, H., Elicone, C., Mark, J., Martin, S. A., and Regnier, F. (1996). Automated analytical system for the examination of protein primary structure. Anal. Chem. 68, 455–462. Huddleston, M. J., Bean, M. F., and Carr, S. A. (1993). Collisional fragmentation of glycopeptides by electrospray ionization LC–MS and LC–MS/MS: Methods for selective detection of glycopeptides in protein digests. Anal. Chem. 65, 877–884. Jacobson, G. R., Schaffer, M. H., Stark, G. R., and Vanaman, T. C. (1983). Specific chemical cleavage in high yield at the amino peptide bonds of cysteine and cystine residues. J. Biol. Chem. 248, 6583–6591. Jemmerson, R., and Paterson, Y. (1986). Mapping epitopes on a protein antigen by the proteolysis of antigen/antibody complexes. Science 232, 1001–1004. Konigsberg, W., and Henderson, H. L. (1983). Removal of sodium dodecyl sulfate from proteins by ion‐pair extraction. Meth. Enzymol. 91, 254–259. Krenkova, J., and Foret, F. (2004). Immobilized microfluidic enzymatic reactors. Electrophoresis 25, 3550–3563. Krogh, T. N., Berg, T., and Hojrup, P. (1999). Protein analysis using enzymes immobilized to paramagnetic beads. Anal. Biochem. 274, 153–162. Ling, V. A., Guzzetta, W., Canova‐Davis, E., Stults, J. T., Hancock, W. S., Covey, T. R., and Shushan, B. I. (1991). Characterization of the tryptic map of recombinant DNA‐derived tissue plasminogen activator by high‐performance liquid chromatography–electrospray ionization mass spectrometry. Anal. Chem. 63, 2909–2915. Mandell, J. G., Baerga‐Ortiz, A., Akashi, S., Takio, K., and Komives, E. A. (2001). Solvent accessibility of the thrombin/thrombomodulin interface. J. Mol. Biol. 306, 575–589.

64

mass spectrometry: modified proteins and glycoconjugates

[3]

Mandell, J. G., Falick, A. M., and Komives, E. A. (1998). Identification of protein/protein interfaces by decreased amide proton solvent accessibility. Proc. Natl. Acad. Sci. USA 95, 14705–14710. Medzihradszky, K. F., Maltby, D. A., Hall, S. C., Settineri, C. A., and Burlingame, A. L. (1994). Characterization of protein N‐glycosylation by reversed‐phase microbore liquid chromatography/electrospray mass spectrometry, complementary mobile phases, and sequential exoglycosidase digestion. J. Am. Soc. Mass Spectrom. 5, 350–358. Medzihradszky, K. F., Darula, Z., Perlson, E., Fainzilber, M., Chalkley, R. J., Ball, H., Greenbaum, D., Bogyo, M., Tyson, D. R., Bradshaw, R. A., and Burlingame, A. L. (2004). O‐sulfonation of serine and threonine: Mass spectrometric detection and characterization of a new post‐translational modification in diverse proteins throughout the eukaryotes. Mol. Cell. Proteomics 3, 429–440. Medzihradszky, K. F., Besman, M. J., and Burlingame, A. L. (1997). Structural characterization of site‐specific N‐glycosylation of recombinant human factor VIII by reversed‐phase high‐performance liquid chromatography‐electrospray ionization mass spectrometry. Anal. Chem. 69, 3986–3994. Medzihradszky, K. F., Gibson, B. W., Kaur, S., Yu, Z., Medzihradszky, D., Burlingame, A. L., and Bass, N. M. (1992). The primary structure of fatty‐acid‐binding protein from nurse shark liver. Structural and evolutionary relationship to the mammalian fatty‐acid‐binding family. Eur. J. Biochem. 203, 327–339. Neville, D. C. A., Rozanas, C. R., Price, E. M., Grius, D. B., Verkman, A. S., and Townsend, R. R. (1997). Evidence for phosphorylation of serine 753 in CFTR using a novel metal‐ion affinity resin and matrix‐assisted laser desorption mass spectrometry. Protein Sci. 6, 2436–2445. Nuwaysir, L. M., and Stults, J. T. (1993). Electrospray ionization mass spectrometry of phosphopeptides isolated by on‐line immobilized metal‐ion affinity chromatography. J. Am. Soc. Mass Spectrom. 4, 662–669. Ostrelund, T., Beussman, D. J., Julenius, K., Poon, P. H., Linse, S., Shabanowitz, J., Hunt, D. F., Schotz, M. C., Derewenda, Z. S., and Holm, C. (1999). Domain identification of hormone‐sensitive lipase by circular dichroism and fluorescence spectroscopy, limited proteolysis, and mass spectrometry. J. Biol. Chem. 274, 15382–15388. Pal, G., Patthy, A., Antal, J., and Graf, L. (2004). Mutant rat trypsin selectively cleaves tyrosyl peptide bonds. Anal. Biochem. 326, 190–199. Posewitz, M. C., and Tempst, P. (1999). Immobilized gallium(III) affinity chromatography of phosphopeptides. Anal. Chem. 71, 2883–2892. Rice, R. H., Means, G. E., and Brown, W. D. (1977). Stabilization of bovine trypsin by reductive methylation. Biochem. Biophys. Acta. 492, 316–321. Riviere, L. R., Fleming, M., Elicone, C., and Tempst, P. (1991). Study and applications of the effects of detergents and chaotropes on enzymatic proteolysis. In ‘‘Techniques in Protein Chemistry II’’ (J. J. Villafranca, ed.), pp. 171–179. Academic Press, San Diego, CA. Rush, R. S., Derby, P. L., Smith, D. M., Merry, C., Rogers, G., Rohde, M. F., and Katta, V. (1995). Microheterogeneity of erythropoietin carbohydrate structure. Anal. Chem. 67, 1442–1452. Rusnak, F., Zhou, J., and Hathaway, G. M. (2002). Identification of phosphorylated and glycosylated sites in peptides by chemically targeted proteolysis. J. Biomol. Tech. 13, 228–237. Schroeder, W. A., Shelton, J. B., and Shelton, J. R. (1969). An examination of conditions for the cleavage of polypeptide chains with cyanogen bromide: Application to catalase. Arch. Biochem. Biophys. 130, 551–556.

[3]

in‐solution digestions for MS

65

Shen, M. L., Johnson, K. L., Mays, D. C., Lipsky, J. L., and Naylor, S. (2001). Determination of in vivo adducts of disulfiram with mitochondrial aldehyde dehydrogenase. Biochem. Pharmacol. 61, 537–545. Sheshberadaran, H., and Payne, L. G. (1988). Protein antigen‐monoclonal antibody contact sites investigated by limited proteolysis of monoclonal antibody‐bound antigen: Protein ‘‘footprinting.’’ Proc. Natl. Acad. Sci. USA 85, 1–5. Simpson, R. J., Moritz, R. L., Nice, E. E., and Grego, B. (1987). A high‐performance liquid chromatography procedure for recovering subnanomole amounts of protein from SDS‐gel electroeluates for gas‐phase sequence analysis. Eur. J. Biochem. 165, 21–29. Smith, B. (1997). Chemical cleavage of polypeptides. In ‘‘Protein Sequencing Protocols. Methods in Molecular Biology’’ (B. J. Smith, ed.), pp. 55–72. Humana Press, Totowa, NJ. Steen, H., Ku¨ ster, B., Fernandez, M., Pandey, A., and Mann, M. (2001). Detection of tyrosine phosphorylated peptides by precursor ion scanning quadrupole TOF mass spectrometry in positive ion mode. Anal. Chem. 73, 1440–1448. Thompson, J. S., Hart, R., Franz, C., Barnouin, K., Ridley, A., and Cramer, A. (2003). Characterization of protein phosphorylation by mass spectrometry using immobilized metal ion affinity chromatography with on‐resin beta‐elimination and Michael addition. Anal. Chem. 75, 3232–3243. Treuheit, M. J., Costello, C. E., and Halsall, H. B. (1992). Analysis of the five glycosylation sites of human alpha 1‐acid glycoprotein. Biochem. J. 283, 105–112. Tschirret‐Guth, R. A., Medzihradszky, K. F., and deMontellano, P. R. O. (1999). Trifluoromethyldiazirinylphenyldiazenes: New hemoprotein active site probes. J. Am. Chem. Soc. 121, 4731–4737. van de Water, J., Deininger, S. O., Macht, M., Przybylski, M., and Gerschwin, M. E. (1997). Detection of molecular determinants and epitope mapping using MALDI–TOF mass spectrometry. Clin. Immunol. Immunpathol. 85, 229–235. Vestling, M. M., Murphy, C. M., and Fenselau, C. (1990). Recognition of trypsin autolysis products by high‐performance liquid chromatography and mass spectrometry. Anal. Chem. 62, 2391–2394. Wakselman, M., and Guibe‐Jampel, E. (1976). 1‐cyano‐4‐dimethylamino‐pyridinium salts— New water‐soluble reagents for cyanylation of protein sulfhydryl groups. J. Chem. Soc. Chem. Commun. 21–22. Wen, D. X., Livingston, B. D., Medzihradszky, K. F., Kelm, S., Burlingame, A. L., and Paulson, J. C. (1992). Primary structure of Gal1,3(4)GlcNAc 2,3‐sialyltransferase reveals a conserved region in the sialyltransferase family. J. Biol. Chem. 267, 21011–21019. Weimbs, T., and Stoffel, W. (1992). Proteolipid protein (PLP) of CNS myelin: Positions of free, disulfide‐bonded, and fatty acid thioester‐linked cysteine residues and implications for the membrane topology of PLP. Biochemistry 31, 12289–12296. Wilm, M., Neubauer, G., and Mann, M. (1996). Parent ion scans of unseparated peptide mixtures. Anal. Chem. 68, 527–533. Wu, J., and Watson, J. T. (1997). A novel methodology for assignment of disulfide bond pairings in proteins. Protein Sci. 6, 391–398. Young, M. M., Tang, N., Hempel, J. C., Oshiro, C. M., Taylor, E. W., Kuntz, I. D., Gibson, B. W., and Dollinger, G. (2000). High throughput protein fold identification by using experimental constraints derived from intramolecular cross‐links and mass spectrometry. Proc. Natl. Acad. Sci. USA 97, 5802–5806. Zhou, W., Merrick, B. A., Khaledi, M. G., and Tomer, K. B. (2000). Detection and sequencing of phosphopeptides affinity bound to immobilized metal ion beads by matrix‐ assisted laser desorption/ionization mass spectrometry. J. Am. Soc. Mass Spectrom. 11, 273–282.